[people/ms/libloc.git] / src / python / location-importer.in

#!/usr/bin/python3
###############################################################################
#                                                                             #
# libloc - A library to determine the location of someone on the Internet     #
#                                                                             #
# Copyright (C) 2020-2021 IPFire Development Team <info@ipfire.org>           #
#                                                                             #
# This library is free software; you can redistribute it and/or               #
# modify it under the terms of the GNU Lesser General Public                  #
# License as published by the Free Software Foundation; either                #
# version 2.1 of the License, or (at your option) any later version.          #
#                                                                             #
# This library is distributed in the hope that it will be useful,             #
# but WITHOUT ANY WARRANTY; without even the implied warranty of              #
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU           #
# Lesser General Public License for more details.                             #
#                                                                             #
###############################################################################

import argparse
import ipaddress
import json
import logging
import math
import re
import socket
import sys
import telnetlib

# Load our location module
import location
import location.database
import location.importer
from location.i18n import _

# Initialise logging
log = logging.getLogger("location.importer")
log.propagate = 1

# Define constants
VALID_ASN_RANGES = (
	(1, 23455),
	(23457, 64495),
	(131072, 4199999999),
)


class CLI(object):
	def parse_cli(self):
		parser = argparse.ArgumentParser(
			description=_("Location Importer Command Line Interface"),
		)
		subparsers = parser.add_subparsers()

		# Global configuration flags
		parser.add_argument("--debug", action="store_true",
			help=_("Enable debug output"))
		parser.add_argument("--quiet", action="store_true",
			help=_("Enable quiet mode"))

		# version
		parser.add_argument("--version", action="version",
			version="%(prog)s @VERSION@")

		# Database
		parser.add_argument("--database-host", required=True,
			help=_("Database Hostname"), metavar=_("HOST"))
		parser.add_argument("--database-name", required=True,
			help=_("Database Name"), metavar=_("NAME"))
		parser.add_argument("--database-username", required=True,
			help=_("Database Username"), metavar=_("USERNAME"))
		parser.add_argument("--database-password", required=True,
			help=_("Database Password"), metavar=_("PASSWORD"))

		# Write Database
		write = subparsers.add_parser("write", help=_("Write database to file"))
		write.set_defaults(func=self.handle_write)
		write.add_argument("file", nargs=1, help=_("Database File"))
		write.add_argument("--signing-key", nargs="?", type=open, help=_("Signing Key"))
		write.add_argument("--backup-signing-key", nargs="?", type=open, help=_("Backup Signing Key"))
		write.add_argument("--vendor", nargs="?", help=_("Sets the vendor"))
		write.add_argument("--description", nargs="?", help=_("Sets a description"))
		write.add_argument("--license", nargs="?", help=_("Sets the license"))
		write.add_argument("--version", type=int, help=_("Database Format Version"))

		# Update WHOIS
		update_whois = subparsers.add_parser("update-whois", help=_("Update WHOIS Information"))
		update_whois.set_defaults(func=self.handle_update_whois)

		# Update announcements
		update_announcements = subparsers.add_parser("update-announcements",
			help=_("Update BGP Annoucements"))
		update_announcements.set_defaults(func=self.handle_update_announcements)
		update_announcements.add_argument("server", nargs=1,
			help=_("Route Server to connect to"), metavar=_("SERVER"))

		# Update overrides
		update_overrides = subparsers.add_parser("update-overrides",
			help=_("Update overrides"),
		)
		update_overrides.add_argument(
			"files", nargs="+", help=_("Files to import"),
		)
		update_overrides.set_defaults(func=self.handle_update_overrides)

		# Import countries
		import_countries = subparsers.add_parser("import-countries",
			help=_("Import countries"),
		)
		import_countries.add_argument("file", nargs=1, type=argparse.FileType("r"),
			help=_("File to import"))
		import_countries.set_defaults(func=self.handle_import_countries)

		args = parser.parse_args()

		# Configure logging
		if args.debug:
			location.logger.set_level(logging.DEBUG)
		elif args.quiet:
			location.logger.set_level(logging.WARNING)

		# Print usage if no action was given
		if not "func" in args:
			parser.print_usage()
			sys.exit(2)

		return args

	def run(self):
		# Parse command line arguments
		args = self.parse_cli()

		# Initialise database
		self.db = self._setup_database(args)

		# Call function
		ret = args.func(args)

		# Return with exit code
		if ret:
			sys.exit(ret)

		# Otherwise just exit
		sys.exit(0)

	def _setup_database(self, ns):
		"""
			Initialise the database
		"""
		# Connect to database
		db = location.database.Connection(
			host=ns.database_host, database=ns.database_name,
			user=ns.database_username, password=ns.database_password,
		)

		with db.transaction():
			db.execute("""
				-- announcements
				CREATE TABLE IF NOT EXISTS announcements(network inet, autnum bigint,
					first_seen_at timestamp without time zone DEFAULT CURRENT_TIMESTAMP,
					last_seen_at timestamp without time zone DEFAULT CURRENT_TIMESTAMP);
				CREATE UNIQUE INDEX IF NOT EXISTS announcements_networks ON announcements(network);
				CREATE INDEX IF NOT EXISTS announcements_family ON announcements(family(network));
				CREATE INDEX IF NOT EXISTS announcements_search ON announcements USING GIST(network inet_ops);

				-- autnums
				CREATE TABLE IF NOT EXISTS autnums(number bigint, name text NOT NULL);
				ALTER TABLE autnums ADD COLUMN IF NOT EXISTS source text;
				CREATE UNIQUE INDEX IF NOT EXISTS autnums_number ON autnums(number);

				-- countries
				CREATE TABLE IF NOT EXISTS countries(
					country_code text NOT NULL, name text NOT NULL, continent_code text NOT NULL);
				CREATE UNIQUE INDEX IF NOT EXISTS countries_country_code ON countries(country_code);

				-- networks
				CREATE TABLE IF NOT EXISTS networks(network inet, country text);
				ALTER TABLE networks ADD COLUMN IF NOT EXISTS original_countries text[];
				ALTER TABLE networks ADD COLUMN IF NOT EXISTS source text;
				CREATE UNIQUE INDEX IF NOT EXISTS networks_network ON networks(network);
				CREATE INDEX IF NOT EXISTS networks_family ON networks USING BTREE(family(network));
				CREATE INDEX IF NOT EXISTS networks_search ON networks USING GIST(network inet_ops);

				-- overrides
				CREATE TABLE IF NOT EXISTS autnum_overrides(
					number bigint NOT NULL,
					name text,
					country text,
					is_anonymous_proxy boolean,
					is_satellite_provider boolean,
					is_anycast boolean
				);
				CREATE UNIQUE INDEX IF NOT EXISTS autnum_overrides_number
					ON autnum_overrides(number);
				ALTER TABLE autnum_overrides ADD COLUMN IF NOT EXISTS source text;
				ALTER TABLE autnum_overrides ADD COLUMN IF NOT EXISTS is_drop boolean;

				CREATE TABLE IF NOT EXISTS network_overrides(
					network inet NOT NULL,
					country text,
					is_anonymous_proxy boolean,
					is_satellite_provider boolean,
					is_anycast boolean
				);
				CREATE UNIQUE INDEX IF NOT EXISTS network_overrides_network
					ON network_overrides(network);
				CREATE INDEX IF NOT EXISTS network_overrides_search
					ON network_overrides USING GIST(network inet_ops);
				ALTER TABLE network_overrides ADD COLUMN IF NOT EXISTS source text;
				ALTER TABLE network_overrides ADD COLUMN IF NOT EXISTS is_drop boolean;
			""")

		return db

	def handle_write(self, ns):
		"""
			Compiles a database in libloc format out of what is in the database
		"""
		# Allocate a writer
		writer = location.Writer(ns.signing_key, ns.backup_signing_key)

		# Set all metadata
		if ns.vendor:
			writer.vendor = ns.vendor

		if ns.description:
			writer.description = ns.description

		if ns.license:
			writer.license = ns.license

		# Add all Autonomous Systems
		log.info("Writing Autonomous Systems...")

		# Select all ASes with a name
		rows = self.db.query("""
			SELECT
				autnums.number AS number,
				COALESCE(
					(SELECT overrides.name FROM autnum_overrides overrides
						WHERE overrides.number = autnums.number),
					autnums.name
				) AS name
				FROM autnums
				WHERE name <> %s ORDER BY number
			""", "")

		for row in rows:
			a = writer.add_as(row.number)
			a.name = row.name

		# Add all networks
		log.info("Writing networks...")

		# Select all known networks
		rows = self.db.query("""
			-- Return a list of those networks enriched with all
			-- other information that we store in the database
			SELECT
				DISTINCT ON (network)
				network,
				autnum,

				-- Country
				COALESCE(
					(
						SELECT country FROM network_overrides overrides
							WHERE networks.network <<= overrides.network
							ORDER BY masklen(overrides.network) DESC
							LIMIT 1
					),
					(
						SELECT country FROM autnum_overrides overrides
							WHERE networks.autnum = overrides.number
					),
					networks.country
				) AS country,

				-- Flags
				COALESCE(
					(
						SELECT is_anonymous_proxy FROM network_overrides overrides
							WHERE networks.network <<= overrides.network
							ORDER BY masklen(overrides.network) DESC
							LIMIT 1
					),
					(
						SELECT is_anonymous_proxy FROM autnum_overrides overrides
							WHERE networks.autnum = overrides.number
					),
					FALSE
				) AS is_anonymous_proxy,
				COALESCE(
					(
						SELECT is_satellite_provider FROM network_overrides overrides
							WHERE networks.network <<= overrides.network
							ORDER BY masklen(overrides.network) DESC
							LIMIT 1
					),
					(
						SELECT is_satellite_provider FROM autnum_overrides overrides
							WHERE networks.autnum = overrides.number
					),
					FALSE
				) AS is_satellite_provider,
				COALESCE(
					(
						SELECT is_anycast FROM network_overrides overrides
							WHERE networks.network <<= overrides.network
							ORDER BY masklen(overrides.network) DESC
							LIMIT 1
					),
					(
						SELECT is_anycast FROM autnum_overrides overrides
							WHERE networks.autnum = overrides.number
					),
					FALSE
				) AS is_anycast,
				COALESCE(
					(
						SELECT is_drop FROM network_overrides overrides
							WHERE networks.network <<= overrides.network
							ORDER BY masklen(overrides.network) DESC
							LIMIT 1
					),
					(
						SELECT is_drop FROM autnum_overrides overrides
							WHERE networks.autnum = overrides.number
					),
					FALSE
				) AS is_drop
			FROM (
				SELECT
					known_networks.network AS network,
					announcements.autnum AS autnum,
					networks.country AS country,

					-- Must be part of returned values for ORDER BY clause
					masklen(announcements.network) AS sort_a,
					masklen(networks.network) AS sort_b
				FROM (
						SELECT network FROM announcements
					UNION ALL
						SELECT network FROM networks
					UNION ALL
						SELECT network FROM network_overrides
					) known_networks
				LEFT JOIN
					announcements ON known_networks.network <<= announcements.network
				LEFT JOIN
					networks ON known_networks.network <<= networks.network
				ORDER BY
					known_networks.network,
					sort_a DESC,
					sort_b DESC
			) networks
		""")

		for row in rows:
			network = writer.add_network(row.network)

			# Save country
			if row.country:
				network.country_code = row.country

			# Save ASN
			if row.autnum:
				network.asn = row.autnum

			# Set flags
			if row.is_anonymous_proxy:
				network.set_flag(location.NETWORK_FLAG_ANONYMOUS_PROXY)

			if row.is_satellite_provider:
				network.set_flag(location.NETWORK_FLAG_SATELLITE_PROVIDER)

			if row.is_anycast:
				network.set_flag(location.NETWORK_FLAG_ANYCAST)

			if row.is_drop:
				network.set_flag(location.NETWORK_FLAG_DROP)

		# Add all countries
		log.info("Writing countries...")
		rows = self.db.query("SELECT * FROM countries ORDER BY country_code")

		for row in rows:
			c = writer.add_country(row.country_code)
			c.continent_code = row.continent_code
			c.name = row.name

		# Write everything to file
		log.info("Writing database to file...")
		for file in ns.file:
			writer.write(file)

	def handle_update_whois(self, ns):
		downloader = location.importer.Downloader()

		# Download all sources
		with self.db.transaction():
			# Create some temporary tables to store parsed data
			self.db.execute("""
				CREATE TEMPORARY TABLE _autnums(number integer NOT NULL, organization text NOT NULL, source text NOT NULL)
					ON COMMIT DROP;
				CREATE UNIQUE INDEX _autnums_number ON _autnums(number);

				CREATE TEMPORARY TABLE _organizations(handle text NOT NULL, name text NOT NULL, source text NOT NULL)
					ON COMMIT DROP;
				CREATE UNIQUE INDEX _organizations_handle ON _organizations(handle);

				CREATE TEMPORARY TABLE _rirdata(network inet NOT NULL, country text NOT NULL, original_countries text[] NOT NULL, source text NOT NULL)
					ON COMMIT DROP;
				CREATE INDEX _rirdata_search ON _rirdata USING BTREE(family(network), masklen(network));
				CREATE UNIQUE INDEX _rirdata_network ON _rirdata(network);
			""")

			# Remove all previously imported content
			self.db.execute("""
				TRUNCATE TABLE networks;
			""")

			# Fetch all valid country codes to check parsed networks aganist...
			rows = self.db.query("SELECT * FROM countries ORDER BY country_code")
			validcountries = []

			for row in rows:
				validcountries.append(row.country_code)

			for source_key in location.importer.WHOIS_SOURCES:
				for single_url in location.importer.WHOIS_SOURCES[source_key]:
					with downloader.request(single_url, return_blocks=True) as f:
						for block in f:
							self._parse_block(block, source_key, validcountries)

			# Process all parsed networks from every RIR we happen to have access to,
			# insert the largest network chunks into the networks table immediately...
			families = self.db.query("SELECT DISTINCT family(network) AS family FROM _rirdata ORDER BY family(network)")

			for family in (row.family for row in families):
				smallest = self.db.get("SELECT MIN(masklen(network)) AS prefix FROM _rirdata WHERE family(network) = %s", family)

				self.db.execute("INSERT INTO networks(network, country, original_countries, source) \
					SELECT network, country, original_countries, source FROM _rirdata WHERE masklen(network) = %s AND family(network) = %s", smallest.prefix, family)

				# ... determine any other prefixes for this network family, ...
				prefixes = self.db.query("SELECT DISTINCT masklen(network) AS prefix FROM _rirdata \
					WHERE family(network) = %s ORDER BY masklen(network) ASC OFFSET 1", family)

				# ... and insert networks with this prefix in case they provide additional
				# information (i. e. subnet of a larger chunk with a different country)
				for prefix in (row.prefix for row in prefixes):
					self.db.execute("""
						WITH candidates AS (
							SELECT
								_rirdata.network,
								_rirdata.country,
								_rirdata.original_countries,
								_rirdata.source
							FROM
								_rirdata
							WHERE
								family(_rirdata.network) = %s
							AND
								masklen(_rirdata.network) = %s
						),
						filtered AS (
							SELECT
								DISTINCT ON (c.network)
								c.network,
								c.country,
								c.original_countries,
								c.source,
								masklen(networks.network),
								networks.country AS parent_country
							FROM
								candidates c
							LEFT JOIN
								networks
							ON
								c.network << networks.network
							ORDER BY
								c.network,
								masklen(networks.network) DESC NULLS LAST
						)
						INSERT INTO
							networks(network, country, original_countries, source)
						SELECT
							network,
							country,
							original_countries,
							source
						FROM
							filtered
						WHERE
							parent_country IS NULL
						OR
							country <> parent_country
						ON CONFLICT DO NOTHING""",
						family, prefix,
					)

			self.db.execute("""
				INSERT INTO autnums(number, name, source)
					SELECT _autnums.number, _organizations.name, _organizations.source FROM _autnums
						JOIN _organizations ON _autnums.organization = _organizations.handle
				ON CONFLICT (number) DO UPDATE SET name = excluded.name;
			""")

		# Download all extended sources
		for source_key in location.importer.EXTENDED_SOURCES:
			for single_url in location.importer.EXTENDED_SOURCES[source_key]:
				with self.db.transaction():
					# Download data
					with downloader.request(single_url) as f:
						for line in f:
							self._parse_line(line, source_key, validcountries)

		# Download and import (technical) AS names from ARIN
		self._import_as_names_from_arin()

	def _check_parsed_network(self, network):
		"""
			Assistive function to detect and subsequently sort out parsed
			networks from RIR data (both Whois and so-called "extended sources"),
			which are or have...

			(a) not globally routable (RFC 1918 space, et al.)
			(b) covering a too large chunk of the IP address space (prefix length
				is < 7 for IPv4 networks, and < 10 for IPv6)
			(c) "0.0.0.0" or "::" as a network address
			(d) are too small for being publicly announced (we have decided not to
				process them at the moment, as they significantly enlarge our
				database without providing very helpful additional information)

			This unfortunately is necessary due to brain-dead clutter across
			various RIR databases, causing mismatches and eventually disruptions.

			We will return False in case a network is not suitable for adding
			it to our database, and True otherwise.
		"""

		if not network or not (isinstance(network, ipaddress.IPv4Network) or isinstance(network, ipaddress.IPv6Network)):
			return False

		if not network.is_global:
			log.debug("Skipping non-globally routable network: %s" % network)
			return False

		if network.version == 4:
			if network.prefixlen < 7:
				log.debug("Skipping too big IP chunk: %s" % network)
				return False

			if network.prefixlen > 24:
				log.debug("Skipping network too small to be publicly announced: %s" % network)
				return False

			if str(network.network_address) == "0.0.0.0":
				log.debug("Skipping network based on 0.0.0.0: %s" % network)
				return False

		elif network.version == 6:
			if network.prefixlen < 10:
				log.debug("Skipping too big IP chunk: %s" % network)
				return False

			if network.prefixlen > 48:
				log.debug("Skipping network too small to be publicly announced: %s" % network)
				return False

			if str(network.network_address) == "::":
				log.debug("Skipping network based on '::': %s" % network)
				return False

		else:
			# This should not happen...
			log.warning("Skipping network of unknown family, this should not happen: %s" % network)
			return False

		# In case we have made it here, the network is considered to
		# be suitable for libloc consumption...
		return True

	def _check_parsed_asn(self, asn):
		"""
			Assistive function to filter Autonomous System Numbers not being suitable
			for adding to our database. Returns False in such cases, and True otherwise.
		"""

		for start, end in VALID_ASN_RANGES:
			if start <= asn and end >= asn:
				return True

		log.info("Supplied ASN %s out of publicly routable ASN ranges" % asn)
		return False

	def _parse_block(self, block, source_key, validcountries = None):
		# Get first line to find out what type of block this is
		line = block[0]

		# aut-num
		if line.startswith("aut-num:"):
			return self._parse_autnum_block(block, source_key)

		# inetnum
		if line.startswith("inet6num:") or line.startswith("inetnum:"):
			return self._parse_inetnum_block(block, source_key, validcountries)

		# organisation
		elif line.startswith("organisation:"):
			return self._parse_org_block(block, source_key)

	def _parse_autnum_block(self, block, source_key):
		autnum = {}
		for line in block:
			# Split line
			key, val = split_line(line)

			if key == "aut-num":
				m = re.match(r"^(AS|as)(\d+)", val)
				if m:
					autnum["asn"] = m.group(2)

			elif key == "org":
				autnum[key] = val.upper()

			elif key == "descr":
				# Save the first description line as well...
				if not key in autnum:
					autnum[key] = val

		# Skip empty objects
		if not autnum or not "asn" in autnum:
			return

		# Insert a dummy organisation handle into our temporary organisations
		# table in case the AS does not have an organisation handle set, but
		# has a description (a quirk often observed in APNIC area), so we can
		# later display at least some string for this AS.
		if not "org" in autnum:
			if "descr" in autnum:
				autnum["org"] = "LIBLOC-%s-ORGHANDLE" % autnum.get("asn")

				self.db.execute("INSERT INTO _organizations(handle, name, source) \
					VALUES(%s, %s, %s) ON CONFLICT (handle) DO NOTHING",
					autnum.get("org"), autnum.get("descr"), source_key,
				)
			else:
				log.warning("ASN %s neither has an organisation handle nor a description line set, omitting" % \
						autnum.get("asn"))
				return

		# Insert into database
		self.db.execute("INSERT INTO _autnums(number, organization, source) \
			VALUES(%s, %s, %s) ON CONFLICT (number) DO UPDATE SET \
				organization = excluded.organization",
			autnum.get("asn"), autnum.get("org"), source_key,
		)

	def _parse_inetnum_block(self, block, source_key, validcountries = None):
		log.debug("Parsing inetnum block:")

		inetnum = {}
		for line in block:
			log.debug(line)

			# Split line
			key, val = split_line(line)

			# Filter any inetnum records which are only referring to IP space
			# not managed by that specific RIR...
			if key == "netname":
				if re.match(r"^(ERX-NETBLOCK|(AFRINIC|ARIN|LACNIC|RIPE)-CIDR-BLOCK|IANA-NETBLOCK-\d{1,3}|NON-RIPE-NCC-MANAGED-ADDRESS-BLOCK|STUB-[\d-]{3,}SLASH\d{1,2})", val.strip()):
					log.debug("Skipping record indicating historic/orphaned data: %s" % val.strip())
					return

			if key == "inetnum":
				start_address, delim, end_address = val.partition("-")

				# Strip any excess space
				start_address, end_address = start_address.rstrip(), end_address.strip()

				# Handle "inetnum" formatting in LACNIC DB (e.g. "24.152.8/22" instead of "24.152.8.0/22")
				if start_address and not (delim or end_address):
					try:
						start_address = ipaddress.ip_network(start_address, strict=False)
					except ValueError:
						start_address = start_address.split("/")
						ldigits = start_address[0].count(".")

						# How many octets do we need to add?
						# (LACNIC does not seem to have a /8 or greater assigned, so the following should suffice.)
						if ldigits == 2:
							start_address = start_address[0] + ".0.0/" + start_address[1]
						elif ldigits == 3:
							start_address = start_address[0] + ".0/" + start_address[1]
						else:
							log.warning("Could not recover IPv4 address from line in LACNIC DB format: %s" % line)
							return

						try:
							start_address = ipaddress.ip_network(start_address, strict=False)
						except ValueError:
							log.warning("Could not parse line in LACNIC DB format: %s" % line)
							return

					# Enumerate first and last IP address of this network
					end_address = start_address[-1]
					start_address = start_address[0]

				else:
					# Convert to IP address
					try:
						start_address = ipaddress.ip_address(start_address)
						end_address   = ipaddress.ip_address(end_address)
					except ValueError:
						log.warning("Could not parse line: %s" % line)
						return

				inetnum["inetnum"] = list(ipaddress.summarize_address_range(start_address, end_address))

			elif key == "inet6num":
				inetnum[key] = [ipaddress.ip_network(val, strict=False)]

			elif key == "country":
				val = val.upper()

				# Catch RIR data objects with more than one country code...
				if not key in inetnum:
					inetnum[key] = []
				else:
					if val in inetnum.get("country"):
						# ... but keep this list distinct...
						continue

				# When people set country codes to "UK", they actually mean "GB"
				if val == "UK":
					val = "GB"

				inetnum[key].append(val)

		# Skip empty objects
		if not inetnum or not "country" in inetnum:
			return

		# Prepare skipping objects with unknown country codes...
		invalidcountries = [singlecountry for singlecountry in inetnum.get("country") if singlecountry not in validcountries]

		# Iterate through all networks enumerated from above, check them for plausibility and insert
		# them into the database, if _check_parsed_network() succeeded
		for single_network in inetnum.get("inet6num") or inetnum.get("inetnum"):
			if self._check_parsed_network(single_network):

				# Skip objects with unknown country codes if they are valid to avoid log spam...
				if validcountries and invalidcountries:
					log.warning("Skipping network with bogus countr(y|ies) %s (original countries: %s): %s" % \
						(invalidcountries, inetnum.get("country"), inetnum.get("inet6num") or inetnum.get("inetnum")))
					break

				# Everything is fine here, run INSERT statement...
				self.db.execute("INSERT INTO _rirdata(network, country, original_countries, source) \
					VALUES(%s, %s, %s, %s) ON CONFLICT (network) DO UPDATE SET country = excluded.country",
					"%s" % single_network, inetnum.get("country")[0], inetnum.get("country"), source_key,
				)

	def _parse_org_block(self, block, source_key):
		org = {}
		for line in block:
			# Split line
			key, val = split_line(line)

			if key == "organisation":
				org[key] = val.upper()
			elif key == "org-name":
				org[key] = val

		# Skip empty objects
		if not org:
			return

		self.db.execute("INSERT INTO _organizations(handle, name, source) \
			VALUES(%s, %s, %s) ON CONFLICT (handle) DO \
			UPDATE SET name = excluded.name",
			org.get("organisation"), org.get("org-name"), source_key,
		)

	def _parse_line(self, line, source_key, validcountries = None):
		# Skip version line
		if line.startswith("2"):
			return

		# Skip comments
		if line.startswith("#"):
			return

		try:
			registry, country_code, type, line = line.split("|", 3)
		except:
			log.warning("Could not parse line: %s" % line)
			return

		# Skip any lines that are for stats only or do not have a country
		# code at all (avoids log spam below)
		if not country_code or country_code == '*':
			return

		# Skip objects with unknown country codes
		if validcountries and country_code not in validcountries:
			log.warning("Skipping line with bogus country '%s': %s" % \
				(country_code, line))
			return

		if type in ("ipv6", "ipv4"):
			return self._parse_ip_line(country_code, type, line, source_key)

	def _parse_ip_line(self, country, type, line, source_key):
		try:
			address, prefix, date, status, organization = line.split("|")
		except ValueError:
			organization = None

			# Try parsing the line without organization
			try:
				address, prefix, date, status = line.split("|")
			except ValueError:
				log.warning("Unhandled line format: %s" % line)
				return

		# Skip anything that isn't properly assigned
		if not status in ("assigned", "allocated"):
			return

		# Cast prefix into an integer
		try:
			prefix = int(prefix)
		except:
			log.warning("Invalid prefix: %s" % prefix)
			return

		# Fix prefix length for IPv4
		if type == "ipv4":
			prefix = 32 - int(math.log(prefix, 2))

		# Try to parse the address
		try:
			network = ipaddress.ip_network("%s/%s" % (address, prefix), strict=False)
		except ValueError:
			log.warning("Invalid IP address: %s" % address)
			return

		if not self._check_parsed_network(network):
			return

		self.db.execute("INSERT INTO networks(network, country, original_countries, source) \
			VALUES(%s, %s, %s, %s) ON CONFLICT (network) DO \
			UPDATE SET country = excluded.country",
			"%s" % network, country, [country], source_key,
		)

	def _import_as_names_from_arin(self):
		downloader = location.importer.Downloader()

		# XXX: Download AS names file from ARIN (note that these names appear to be quite
		# technical, not intended for human consumption, as description fields in
		# organisation handles for other RIRs are - however, this is what we have got,
		# and in some cases, it might be still better than nothing)
		with downloader.request("https://ftp.arin.net/info/asn.txt", return_blocks=False) as f:
			for line in f:
				# Convert binary line to string...
				line = str(line)

				# ... valid lines start with a space, followed by the number of the Autonomous System ...
				if not line.startswith(" "):
					continue

				# Split line and check if there is a valid ASN in it...
				asn, name = line.split()[0:2]

				try:
					asn = int(asn)
				except ValueError:
					log.debug("Skipping ARIN AS names line not containing an integer for ASN")
					continue

				# Filter invalid ASNs...
				if not self._check_parsed_asn(asn):
					continue

				# Skip any AS name that appears to be a placeholder for a different RIR or entity...
				if re.match(r"^(ASN-BLK|)(AFCONC|AFRINIC|APNIC|ASNBLK|DNIC|LACNIC|RIPE|IANA)(?:\d?$|\-)", name):
					continue

				# Bail out in case the AS name contains anything we do not expect here...
				if re.search(r"[^a-zA-Z0-9-_]", name):
					log.debug("Skipping ARIN AS name for %s containing invalid characters: %s" % \
							(asn, name))

				# Things look good here, run INSERT statement and skip this one if we already have
				# a (better?) name for this Autonomous System...
				self.db.execute("""
					INSERT INTO autnums(
						number,
						name,
						source
					) VALUES (%s, %s, %s)
					ON CONFLICT (number) DO NOTHING""",
					asn,
					name,
					"ARIN",
				)

	def handle_update_announcements(self, ns):
		server = ns.server[0]

		with self.db.transaction():
			if server.startswith("/"):
				self._handle_update_announcements_from_bird(server)
			else:
				self._handle_update_announcements_from_telnet(server)

			# Purge anything we never want here
			self.db.execute("""
				-- Delete default routes
				DELETE FROM announcements WHERE network = '::/0' OR network = '0.0.0.0/0';

				-- Delete anything that is not global unicast address space
				DELETE FROM announcements WHERE family(network) = 6 AND NOT network <<= '2000::/3';

				-- DELETE "current network" address space
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '0.0.0.0/8';

				-- DELETE local loopback address space
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '127.0.0.0/8';

				-- DELETE RFC 1918 address space
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '10.0.0.0/8';
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '172.16.0.0/12';
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '192.168.0.0/16';

				-- DELETE test, benchmark and documentation address space
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '192.0.0.0/24';
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '192.0.2.0/24';
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '198.18.0.0/15';
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '198.51.100.0/24';
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '203.0.113.0/24';

				-- DELETE CGNAT address space (RFC 6598)
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '100.64.0.0/10';

				-- DELETE link local address space
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '169.254.0.0/16';

				-- DELETE IPv6 to IPv4 (6to4) address space (RFC 3068)
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '192.88.99.0/24';
				DELETE FROM announcements WHERE family(network) = 6 AND network <<= '2002::/16';

				-- DELETE multicast and reserved address space
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '224.0.0.0/4';
				DELETE FROM announcements WHERE family(network) = 4 AND network <<= '240.0.0.0/4';

				-- Delete networks that are too small to be in the global routing table
				DELETE FROM announcements WHERE family(network) = 6 AND masklen(network) > 48;
				DELETE FROM announcements WHERE family(network) = 4 AND masklen(network) > 24;

				-- Delete any non-public or reserved ASNs
				DELETE FROM announcements WHERE NOT (
					(autnum >= 1 AND autnum <= 23455)
					OR
					(autnum >= 23457 AND autnum <= 64495)
					OR
					(autnum >= 131072 AND autnum <= 4199999999)
				);

				-- Delete everything that we have not seen for 14 days
				DELETE FROM announcements WHERE last_seen_at <= CURRENT_TIMESTAMP - INTERVAL '14 days';
			""")

	def _handle_update_announcements_from_bird(self, server):
		# Pre-compile the regular expression for faster searching
		route = re.compile(b"^\s(.+?)\s+.+?\[AS(.*?).\]$")

		log.info("Requesting routing table from Bird (%s)" % server)

		# Send command to list all routes
		for line in self._bird_cmd(server, "show route"):
			m = route.match(line)
			if not m:
				log.debug("Could not parse line: %s" % line.decode())
				continue

			# Fetch the extracted network and ASN
			network, autnum = m.groups()

			# Insert it into the database
			self.db.execute("INSERT INTO announcements(network, autnum) \
				VALUES(%s, %s) ON CONFLICT (network) DO \
				UPDATE SET autnum = excluded.autnum, last_seen_at = CURRENT_TIMESTAMP",
				network.decode(), autnum.decode(),
			)

	def _handle_update_announcements_from_telnet(self, server):
		# Pre-compile regular expression for routes
		route = re.compile(b"^\*[\s\>]i([^\s]+).+?(\d+)\si\r\n", re.MULTILINE|re.DOTALL)

		with telnetlib.Telnet(server) as t:
			# Enable debug mode
			#if ns.debug:
			#	t.set_debuglevel(10)

			# Wait for console greeting
			greeting = t.read_until(b"> ", timeout=30)
			if not greeting:
				log.error("Could not get a console prompt")
				return 1

			# Disable pagination
			t.write(b"terminal length 0\n")

			# Wait for the prompt to return
			t.read_until(b"> ")

			# Fetch the routing tables
			for protocol in ("ipv6", "ipv4"):
				log.info("Requesting %s routing table" % protocol)

				# Request the full unicast routing table
				t.write(b"show bgp %s unicast\n" % protocol.encode())

				# Read entire header which ends with "Path"
				t.read_until(b"Path\r\n")

				while True:
					# Try reading a full entry
					# Those might be broken across multiple lines but ends with i
					line = t.read_until(b"i\r\n", timeout=5)
					if not line:
						break

					# Show line for debugging
					#log.debug(repr(line))

					# Try finding a route in here
					m = route.match(line)
					if m:
						network, autnum = m.groups()

						# Convert network to string
						network = network.decode()

						# Append /24 for IPv4 addresses
						if not "/" in network and not ":" in network:
							network = "%s/24" % network

						# Convert AS number to integer
						autnum = int(autnum)

						log.info("Found announcement for %s by %s" % (network, autnum))

						self.db.execute("INSERT INTO announcements(network, autnum) \
							VALUES(%s, %s) ON CONFLICT (network) DO \
							UPDATE SET autnum = excluded.autnum, last_seen_at = CURRENT_TIMESTAMP",
							network, autnum,
						)

				log.info("Finished reading the %s routing table" % protocol)

	def _bird_cmd(self, socket_path, command):
		# Connect to the socket
		s = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
		s.connect(socket_path)

		# Allocate some buffer
		buffer = b""

		# Send the command
		s.send(b"%s\n" % command.encode())

		while True:
			# Fill up the buffer
			buffer += s.recv(4096)

			while True:
				# Search for the next newline
				pos = buffer.find(b"\n")

				# If we cannot find one, we go back and read more data
				if pos <= 0:
					break

				# Cut after the newline character
				pos += 1

				# Split the line we want and keep the rest in buffer
				line, buffer = buffer[:pos], buffer[pos:]

				# Look for the end-of-output indicator
				if line == b"0000 \n":
					return

				# Otherwise return the line
				yield line

	def handle_update_overrides(self, ns):
		with self.db.transaction():
			# Drop all data that we have
			self.db.execute("""
				TRUNCATE TABLE autnum_overrides;
				TRUNCATE TABLE network_overrides;
			""")

			# Update overrides for various cloud providers big enough to publish their own IP
			# network allocation lists in a machine-readable format...
			self._update_overrides_for_aws()

			# Update overrides for Spamhaus DROP feeds...
			self._update_overrides_for_spamhaus_drop()

			for file in ns.files:
				log.info("Reading %s..." % file)

				with open(file, "rb") as f:
					for type, block in location.importer.read_blocks(f):
						if type == "net":
							network = block.get("net")
							# Try to parse and normalise the network
							try:
								network = ipaddress.ip_network(network, strict=False)
							except ValueError as e:
								log.warning("Invalid IP network: %s: %s" % (network, e))
								continue

							# Prevent that we overwrite all networks
							if network.prefixlen == 0:
								log.warning("Skipping %s: You cannot overwrite default" % network)
								continue

							self.db.execute("""
								INSERT INTO network_overrides(
									network,
									country,
									source,
									is_anonymous_proxy,
									is_satellite_provider,
									is_anycast,
									is_drop
								) VALUES (%s, %s, %s, %s, %s, %s, %s)
								ON CONFLICT (network) DO NOTHING""",
								"%s" % network,
								block.get("country"),
								"manual",
								self._parse_bool(block, "is-anonymous-proxy"),
								self._parse_bool(block, "is-satellite-provider"),
								self._parse_bool(block, "is-anycast"),
								self._parse_bool(block, "drop"),
							)

						elif type == "aut-num":
							autnum = block.get("aut-num")

							# Check if AS number begins with "AS"
							if not autnum.startswith("AS"):
								log.warning("Invalid AS number: %s" % autnum)
								continue

							# Strip "AS"
							autnum = autnum[2:]

							self.db.execute("""
								INSERT INTO autnum_overrides(
									number,
									name,
									country,
									source,
									is_anonymous_proxy,
									is_satellite_provider,
									is_anycast,
									is_drop
								) VALUES(%s, %s, %s, %s, %s, %s, %s, %s)
								ON CONFLICT DO NOTHING""",
								autnum,
								block.get("name"),
								block.get("country"),
								"manual",
								self._parse_bool(block, "is-anonymous-proxy"),
								self._parse_bool(block, "is-satellite-provider"),
								self._parse_bool(block, "is-anycast"),
								self._parse_bool(block, "drop"),
							)

						else:
							log.warning("Unsupported type: %s" % type)

	def _update_overrides_for_aws(self):
		# Download Amazon AWS IP allocation file to create overrides...
		downloader = location.importer.Downloader()

		try:
			with downloader.request("https://ip-ranges.amazonaws.com/ip-ranges.json", return_blocks=False) as f:
				aws_ip_dump = json.load(f.body)
		except Exception as e:
			log.error("unable to preprocess Amazon AWS IP ranges: %s" % e)
			return

		# XXX: Set up a dictionary for mapping a region name to a country. Unfortunately,
		# there seems to be no machine-readable version available of this other than
		# https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-regions-availability-zones.html
		# (worse, it seems to be incomplete :-/ ); https://www.cloudping.cloud/endpoints
		# was helpful here as well.
		aws_region_country_map = {
				"af-south-1": "ZA",
				"ap-east-1": "HK",
				"ap-south-1": "IN",
				"ap-south-2": "IN",
				"ap-northeast-3": "JP",
				"ap-northeast-2": "KR",
				"ap-southeast-1": "SG",
				"ap-southeast-2": "AU",
				"ap-southeast-3": "MY",
				"ap-southeast-4": "AU",
				"ap-northeast-1": "JP",
				"ca-central-1": "CA",
				"eu-central-1": "DE",
				"eu-central-2": "CH",
				"eu-west-1": "IE",
				"eu-west-2": "GB",
				"eu-south-1": "IT",
				"eu-south-2": "ES",
				"eu-west-3": "FR",
				"eu-north-1": "SE",
				"il-central-1": "IL", # XXX: This one is not documented anywhere except for ip-ranges.json itself
				"me-central-1": "AE",
				"me-south-1": "BH",
				"sa-east-1": "BR"
				}

		# Fetch all valid country codes to check parsed networks aganist...
		rows = self.db.query("SELECT * FROM countries ORDER BY country_code")
		validcountries = []

		for row in rows:
			validcountries.append(row.country_code)

		with self.db.transaction():
			for snetwork in aws_ip_dump["prefixes"] + aws_ip_dump["ipv6_prefixes"]:
				try:
					network = ipaddress.ip_network(snetwork.get("ip_prefix") or snetwork.get("ipv6_prefix"), strict=False)
				except ValueError:
					log.warning("Unable to parse line: %s" % snetwork)
					continue

				# Sanitize parsed networks...
				if not self._check_parsed_network(network):
					continue

				# Determine region of this network...
				region = snetwork["region"]
				cc = None
				is_anycast = False

				# Any region name starting with "us-" will get "US" country code assigned straight away...
				if region.startswith("us-"):
					cc = "US"
				elif region.startswith("cn-"):
					# ... same goes for China ...
					cc = "CN"
				elif region == "GLOBAL":
					# ... funny region name for anycast-like networks ...
					is_anycast = True
				elif region in aws_region_country_map:
					# ... assign looked up country code otherwise ...
					cc = aws_region_country_map[region]
				else:
					# ... and bail out if we are missing something here
					log.warning("Unable to determine country code for line: %s" % snetwork)
					continue

				# Skip networks with unknown country codes
				if not is_anycast and validcountries and cc not in validcountries:
					log.warning("Skipping Amazon AWS network with bogus country '%s': %s" % \
						(cc, network))
					return

				# Conduct SQL statement...
				self.db.execute("""
					INSERT INTO network_overrides(
						network,
						country,
						source,
						is_anonymous_proxy,
						is_satellite_provider,
						is_anycast
					) VALUES (%s, %s, %s, %s, %s, %s)
					ON CONFLICT (network) DO NOTHING""",
					"%s" % network,
					cc,
					"Amazon AWS IP feed",
					None,
					None,
					is_anycast,
				)


	def _update_overrides_for_spamhaus_drop(self):
		downloader = location.importer.Downloader()

		ip_urls = [
					"https://www.spamhaus.org/drop/drop.txt",
					"https://www.spamhaus.org/drop/edrop.txt",
					"https://www.spamhaus.org/drop/dropv6.txt"
				]

		asn_urls = [
					"https://www.spamhaus.org/drop/asndrop.txt"
				]

		for url in ip_urls:
			try:
				with downloader.request(url, return_blocks=False) as f:
					fcontent = f.body.readlines()
			except Exception as e:
				log.error("Unable to download Spamhaus DROP URL %s: %s" % (url, e))
				return

			# Iterate through every line, filter comments and add remaining networks to
			# the override table in case they are valid...
			with self.db.transaction():
				for sline in fcontent:

					# The response is assumed to be encoded in UTF-8...
					sline = sline.decode("utf-8")

					# Comments start with a semicolon...
					if sline.startswith(";"):
						continue

					# Extract network and ignore anything afterwards...
					try:
						network = ipaddress.ip_network(sline.split()[0], strict=False)
					except ValueError:
						log.error("Unable to parse line: %s" % sline)
						continue

					# Sanitize parsed networks...
					if not self._check_parsed_network(network):
						log.warning("Skipping bogus network found in Spamhaus DROP URL %s: %s" % \
							(url, network))
						continue

					# Conduct SQL statement...
					self.db.execute("""
						INSERT INTO network_overrides(
							network,
							source,
							is_drop
						) VALUES (%s, %s, %s)
						ON CONFLICT (network) DO UPDATE SET is_drop = True""",
						"%s" % network,
						"Spamhaus DROP lists",
						True
					)

		for url in asn_urls:
			try:
				with downloader.request(url, return_blocks=False) as f:
					fcontent = f.body.readlines()
			except Exception as e:
				log.error("Unable to download Spamhaus DROP URL %s: %s" % (url, e))
				return

			# Iterate through every line, filter comments and add remaining ASNs to
			# the override table in case they are valid...
			with self.db.transaction():
				for sline in fcontent:

					# The response is assumed to be encoded in UTF-8...
					sline = sline.decode("utf-8")

					# Comments start with a semicolon...
					if sline.startswith(";"):
						continue

					# Throw away anything after the first space...
					sline = sline.split()[0]

					# ... strip the "AS" prefix from it ...
					sline = sline.strip("AS")

					# ... and convert it into an integer. Voila.
					asn = int(sline)

					# Filter invalid ASNs...
					if not self._check_parsed_asn(asn):
						log.warning("Skipping bogus ASN found in Spamhaus DROP URL %s: %s" % \
							(url, asn))
						continue

					# Conduct SQL statement...
					self.db.execute("""
						INSERT INTO autnum_overrides(
							number,
							source,
							is_drop
						) VALUES (%s, %s, %s)
						ON CONFLICT (number) DO UPDATE SET is_drop = True""",
						"%s" % asn,
						"Spamhaus ASN-DROP list",
						True
					)

	@staticmethod
	def _parse_bool(block, key):
		val = block.get(key)

		# There is no point to proceed when we got None
		if val is None:
			return

		# Convert to lowercase
		val = val.lower()

		# True
		if val in ("yes", "1"):
			return True

		# False
		if val in ("no", "0"):
			return False

		# Default to None
		return None

	def handle_import_countries(self, ns):
		with self.db.transaction():
			# Drop all data that we have
			self.db.execute("TRUNCATE TABLE countries")

			for file in ns.file:
				for line in file:
					line = line.rstrip()

					# Ignore any comments
					if line.startswith("#"):
						continue

					try:
						country_code, continent_code, name = line.split(maxsplit=2)
					except:
						log.warning("Could not parse line: %s" % line)
						continue

					self.db.execute("INSERT INTO countries(country_code, name, continent_code) \
						VALUES(%s, %s, %s) ON CONFLICT DO NOTHING", country_code, name, continent_code)


def split_line(line):
	key, colon, val = line.partition(":")

	# Strip any excess space
	key = key.strip()
	val = val.strip()

	return key, val

def main():
	# Run the command line interface
	c = CLI()
	c.run()

main()